{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:O5U6WR22GCOGXXZLHTIOPUTMR4","short_pith_number":"pith:O5U6WR22","schema_version":"1.0","canonical_sha256":"7769eb475a309c6bdf2b3cd0e7d26c8f087508e2812c7c56742264e85a109b77","source":{"kind":"arxiv","id":"2504.10462","version":1},"attestation_state":"computed","paper":{"title":"The Scalability of Simplicity: Empirical Analysis of Vision-Language Learning with a Single Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haochen Wang, Jiacong Wang, Jiashi Feng, Jun Hao Liew, Weixian Lei, Xiangtai Li, Zilong Huang","submitted_at":"2025-04-14T17:50:20Z","abstract_excerpt":"This paper introduces SAIL, a single transformer unified multimodal large language model (MLLM) that integrates raw pixel encoding and language decoding within a singular architecture. Unlike existing modular MLLMs, which rely on a pre-trained vision transformer (ViT), SAIL eliminates the need for a separate vision encoder, presenting a more minimalist architecture design. Instead of introducing novel architectural components, SAIL adapts mix-attention mechanisms and multimodal positional encodings to better align with the distinct characteristics of visual and textual modalities. We systemati"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2504.10462","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-14T17:50:20Z","cross_cats_sorted":[],"title_canon_sha256":"30adc8979bed256356a6b86c926dd1c3c7ade2ce12dd99f6e610020eb8e6974a","abstract_canon_sha256":"255859b11957126f10750d6b5974b419f7b48c949c5b978b935bee55428759be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:00.867204Z","signature_b64":"fAd5WoxzSdPGlImG2ZX/8pKme2RXkGnsLsPKAp943hRfXR7kBT6ENlvFSmZWxMaja81HFi16no0Kqm/VNHM7DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7769eb475a309c6bdf2b3cd0e7d26c8f087508e2812c7c56742264e85a109b77","last_reissued_at":"2026-07-05T10:49:00.866717Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:00.866717Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Scalability of Simplicity: Empirical Analysis of Vision-Language Learning with a Single Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haochen Wang, Jiacong Wang, Jiashi Feng, Jun Hao Liew, Weixian Lei, Xiangtai Li, Zilong Huang","submitted_at":"2025-04-14T17:50:20Z","abstract_excerpt":"This paper introduces SAIL, a single transformer unified multimodal large language model (MLLM) that integrates raw pixel encoding and language decoding within a singular architecture. Unlike existing modular MLLMs, which rely on a pre-trained vision transformer (ViT), SAIL eliminates the need for a separate vision encoder, presenting a more minimalist architecture design. Instead of introducing novel architectural components, SAIL adapts mix-attention mechanisms and multimodal positional encodings to better align with the distinct characteristics of visual and textual modalities. We systemati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10462","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.10462/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.10462","created_at":"2026-07-05T10:49:00.866777+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.10462v1","created_at":"2026-07-05T10:49:00.866777+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10462","created_at":"2026-07-05T10:49:00.866777+00:00"},{"alias_kind":"pith_short_12","alias_value":"O5U6WR22GCOG","created_at":"2026-07-05T10:49:00.866777+00:00"},{"alias_kind":"pith_short_16","alias_value":"O5U6WR22GCOGXXZL","created_at":"2026-07-05T10:49:00.866777+00:00"},{"alias_kind":"pith_short_8","alias_value":"O5U6WR22","created_at":"2026-07-05T10:49:00.866777+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.28820","citing_title":"From Pixels to Words -- Towards Native One-Vision Models at Scale","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2505.16416","citing_title":"Circle-RoPE: Cone-like Decoupled Rotary Positional Embedding for Large Vision-Language Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19852","citing_title":"Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2506.11991","citing_title":"VGR: Visual Grounded Reasoning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13080","citing_title":"Learning to See What You Need: Gaze Attention for Multimodal Large Language Models","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4","json":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4.json","graph_json":"https://pith.science/api/pith-number/O5U6WR22GCOGXXZLHTIOPUTMR4/graph.json","events_json":"https://pith.science/api/pith-number/O5U6WR22GCOGXXZLHTIOPUTMR4/events.json","paper":"https://pith.science/paper/O5U6WR22"},"agent_actions":{"view_html":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4","download_json":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4.json","view_paper":"https://pith.science/paper/O5U6WR22","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.10462&json=true","fetch_graph":"https://pith.science/api/pith-number/O5U6WR22GCOGXXZLHTIOPUTMR4/graph.json","fetch_events":"https://pith.science/api/pith-number/O5U6WR22GCOGXXZLHTIOPUTMR4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4/action/storage_attestation","attest_author":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4/action/author_attestation","sign_citation":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4/action/citation_signature","submit_replication":"https://pith.science/pith/O5U6WR22GCOGXXZLHTIOPUTMR4/action/replication_record"}},"created_at":"2026-07-05T10:49:00.866777+00:00","updated_at":"2026-07-05T10:49:00.866777+00:00"}